0
$\begingroup$

I am using SVM and Random forest for classification purpose on a dataset.

I am able to optimise the SVM parameters and SVM is providing very good performance in terms of accuracy, recall.

But,at the same time Random Forest (RF) is not providing good perforamnce on the same datset. There is big diference between the two classifier's performance. I have tried many permutations combinatiosn for RF configuration, but its perfroamnce is not improved.

Please suggest what should I try further for RF.

$\endgroup$
0
$\begingroup$

RF doesn't always outperform SVM. Same goes for any two ML algorithms. It depends on the relation between the data and the target. You can try GBM if you must use trees, but that requires more hyper-parameter tuning.

Read this and see if your problem satisfies any of the criteria.

$\endgroup$
  • $\begingroup$ But, some how I have to use RF for my work. Could you suggest, what parameter I should configure, so that it works properly. I find RF is biased towards the larger size category of class. FYI, I am using rapidminer tool for experimentation. Thanks $\endgroup$ – kailash Jan 18 '16 at 11:20
  • $\begingroup$ What parameters are in your control? Is it a multiclass problem? I am not aware of that tool. $\endgroup$ – Ujjwal Kumar Jan 18 '16 at 18:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.